Search results for: bayesian inference

Authors: Alexandre Leclerc, Christian Gout, Carole Le Guyader, Pierre Besset, Carlos Miranda, Olivier Gibaru

Abstract:

In this work, we address the challenges faced by human operators in visually inspecting production lines in factories, especially in automotive manufacturing involving numerous checkpoints to be handled every minute for extended hours. This tedious and error-prone process calls for a solution, and we propose an AI-powered assistant to achieve this goal. The objective is to design a system that can efficiently perform visual inspections on moving items and assemblies, in a structured manner, such as on conveyors and production lines. The system task is to visually verify specific part conformity for each assembly or class of assembly. For instance, in an automotive assembly line, the assistant may need to check the presence of headlights and ensure proper rim mounting. To meet these requirements, the assistant must detect the passage of relevant assemblies and perform anomaly detection on relevent parts. Notably, the system copes with the a prior knowledge deficit challenge of which assemblies and parts the users will choose for inspection. Therefore, it needs to be versatile enough to work across a wide range of industrial use cases. By incorporating these considerations, the proposed AI-powered assistant aims to reduce the difficulty, tedium and unreliability of human visual checks on production lines. Constraints specific to the industrial sector underpin the proposal: • Easy and fast setup (< 1 hour). • Strong hardware constraints on GPU power and memory. • Short inference time (< 100 ms). The proposed approach involves leveraging recent advances in mathematical models and computer science tools complying with the above requirements. It can be summarized as follows: 1) Implementing state-of-the-art deep neural networks trained on vast and diverse datasets (known as foundation models) to assist with annotation, facilitating rapid preparation for detector training. 2) Utilizing the latest models capable of handling object detection and instance segmentation tasks to track the objects needing analysis. Result refinement is then achieved through trajectory analysis of the examined areas. 3) Incorporating a pre-trained classifier and fine-tuning it using detector-acquired data. 4) Inferring the classification model on the detections obtained from the trackers to warn the operator quickly when necessary. Furthermore, we subjected these methods to continuous testing in real industrial environments to ensure their practical applicability and effectiveness.

Keywords: anomaly detection, computer vision, deep learning, image and video processing, image segmentation, industrial inspection, production line

Procedia PDF Downloads 52

Authors: Maximilian Engelhardt, Stephan Seeck

Abstract:

The COVID-19 pandemic is accelerating the already growing field of e-commerce. The resulting urban freight transport volume leads to traffic and negative environmental impact. Furthermore, the service level of parcel logistics service provider is lacking far behind the expectations of consumer. These challenges can be solved by radically reorganize the urban last mile distribution structure: parcels could be consolidated in a micro hub within the inner city and delivered within time windows by cargo bike. This approach leads to a significant improvement of consumer satisfaction with their overall delivery experience. However, this approach also leads to significantly increased costs per parcel. While there is a relevant share of online shoppers that are willing to pay for such a delivery service there are no deeper insights about this target group available in the literature. Being aware of the importance of knowing target groups for businesses, the aim of this paper is to elaborate the most important factors that determine the willingness to pay for sustainable and service-oriented parcel delivery (factor analysis) and to derive customer segments (cluster analysis). In order to answer those questions, a data set is analyzed using quantitative methods of multivariate statistics. The data set was generated via an online survey in September and October 2020 within the five largest cities in Germany (n = 1.071). The data set contains socio-demographic, living-related and value-related variables, e.g. age, income, city, living situation and willingness to pay. In a prior work of the author, the data was analyzed applying descriptive and inference statistical methods that only provided limited insights regarding the above-mentioned research questions. The analysis in an exploratory way using factor and cluster analysis promise deeper insights of relevant influencing factors and segments for user behavior of the mentioned parcel delivery concept. The analysis model is built and implemented with help of the statistical software language R. The data analysis is currently performed and will be completed in December 2021. It is expected that the results will show the most relevant factors that are determining user behavior of sustainable and service-oriented parcel deliveries (e.g. age, current service experience, willingness to pay) and give deeper insights in characteristics that describe the segments that are more or less willing to pay for a better parcel delivery service. Based on the expected results, relevant implications and conclusions can be derived for startups that are about to change the way parcels are delivered: more customer-orientated by time window-delivery and parcel consolidation, more environmental-friendly by cargo bike. The results will give detailed insights regarding their target groups of parcel recipients. Further research can be conducted by exploring alternative revenue models (beyond the parcel recipient) that could compensate the additional costs, e.g. online-shops that increase their service-level or municipalities that reduce traffic on their streets.

Keywords: customer segmentation, e-commerce, last mile delivery, parcel service, urban logistics, willingness-to-pay

Procedia PDF Downloads 81

Authors: Andrew N. Saylor, James R. Peters

Abstract:

Scoliosis is a complex 3D deformity of the thoracic and lumbar spines, clinically diagnosed by measurement of a Cobb angle of 10 degrees or more on a coronal X-ray. The Cobb angle is the angle made by the lines drawn along the proximal and distal endplates of the respective proximal and distal vertebrae comprising the curve. Traditionally, Cobb angles are measured manually using either a marker, straight edge, and protractor or image measurement software. The task of measuring the Cobb angle can also be represented by a function taking the spine geometry rendered using X-ray imaging as input and returning the approximate angle. Although the form of such a function may be unknown, it can be approximated using artificial neural networks (ANNs). The performance of ANNs is affected by many factors, including the choice of activation function and network architecture; however, the effects of these parameters on the accuracy of scoliotic deformity measurements are poorly understood. Therefore, the objective of this study was to systematically investigate the effect of ANN architecture and activation function on Cobb angle measurement from the coronal X-rays of scoliotic subjects. The data set for this study consisted of 609 coronal chest X-rays of scoliotic subjects divided into 481 training images and 128 test images. These data, which included labeled Cobb angle measurements, were obtained from the SpineWeb online database. In order to normalize the input data, each image was resized using bi-linear interpolation to a size of 500 × 187 pixels, and the pixel intensities were scaled to be between 0 and 1. A fully connected (dense) ANN with a fixed cost function (mean squared error), batch size (10), and learning rate (0.01) was developed using Python Version 3.7.3 and TensorFlow 1.13.1. The activation functions (sigmoid, hyperbolic tangent [tanh], or rectified linear units [ReLU]), number of hidden layers (1, 3, 5, or 10), and number of neurons per layer (10, 100, or 1000) were varied systematically to generate a total of 36 network conditions. Stochastic gradient descent with early stopping was used to train each network. Three trials were run per condition, and the final mean squared errors and mean absolute errors were averaged to quantify the network response for each condition. The network that performed the best used ReLU neurons had three hidden layers, and 100 neurons per layer. The average mean squared error of this network was 222.28 ± 30 degrees2, and the average mean absolute error was 11.96 ± 0.64 degrees. It is also notable that while most of the networks performed similarly, the networks using ReLU neurons, 10 hidden layers, and 1000 neurons per layer, and those using Tanh neurons, one hidden layer, and 10 neurons per layer performed markedly worse with average mean squared errors greater than 400 degrees2 and average mean absolute errors greater than 16 degrees. From the results of this study, it can be seen that the choice of ANN architecture and activation function has a clear impact on Cobb angle inference from coronal X-rays of scoliotic subjects.

Keywords: scoliosis, artificial neural networks, cobb angle, medical imaging

Procedia PDF Downloads 95

Authors: Sonia P. Marrero Duran, Lilian Noya Dominguez, Lisandra Quintana Alvarez, Evert Martinez Perez, Ana Julia Acevedo Urquiaga

Abstract:

Cuban business and economic policy are in the constant update as well as facing a client ever more knowledgeable and demanding. For that reason become fundamental for companies competitiveness through the optimization of its processes and services. One of the Cuban’s pillars, which has been sustained since the triumph of the Cuban Revolution back in 1959, is the free health service to all those who need it. This service is offered without any charge under the concept of preserving human life, but it implied costly management processes and logistics services to be able to supply the necessary medicines to all the units who provide health services. One of the key actors on the medicine supply chain is the Havana Distribution Center (HDC), which is responsible for the delivery of medicines in the province; as well as the acquisition of medicines from national and international producers and its subsequent transport to health care units and pharmacies in time, and with the required quality. This HDC also carries for all distribution centers in the country. Given the eminent need to create an actor in the supply chain that specializes in the medicines supply, the possibility of centralizing this operation in a logistics service provider is analyzed. Based on this decision, pharmacies operate as clients of the logistic service center whose main function is to centralize all logistics operations associated with the medicine supply chain. The HDC is precisely the logistic service provider in Havana and it is the center of this research. In 2017 the pharmacies had affectations in the availability of medicine due to deficiencies in the distribution routes. This is caused by the fact that they are not based on routing studies, besides the long distribution cycle. The distribution routs are fixed, attend only one type of customer and there respond to a territorial location by the municipality. Taking into consideration the above-mentioned problem, the objective of this research is to optimize the routes system in the Havana Distribution Center. To accomplish this objective, the techniques applied were document analysis, random sampling, statistical inference and tools such as Ishikawa diagram and the computerized software’s: ArcGis, Osmand y MapIfnfo. As a result, were analyzed four distribution alternatives; the actual rout, by customer type, by the municipality and the combination of the two last. It was demonstrated that the territorial location alternative does not take full advantage of the transportation capacities or the distance of the trips, which leads to elevated costs breaking whit the current ways of distribution and the currents characteristics of the clients. The principal finding of the investigation was the optimum option distribution rout is the 4th one that is formed by hospitals and the join of pharmacies, stomatology clinics, polyclinics and maternal and elderly homes. This solution breaks the territorial location by the municipality and permits different distribution cycles in dependence of medicine consumption and transport availability.

Keywords: computerized geographic software, distribution, distribution routs, vehicle problem routing (VPR)

Procedia PDF Downloads 132

Authors: Jonardan Koner, Avinash Purandare

Abstract:

In today’s globalized world international business is a very key area for the country's growth. Some of the strategic areas for holding up a country’s international business to grow are in the areas of connecting Ports, Road Network, and Rail Network. India’s International Business is booming both in Exports as well as Imports. Ports play a very central part in the growth of international trade and ensuring competitive ports is of critical importance. India has a long coastline which is a big asset for the country as it has given the opportunity for development of a large number of major and minor ports which will contribute to the maritime trades’ development. The National Economic Development of India requires a well-functioning seaport system. To know the comparative strength of Indian ports over South-east Asian similar ports, the study is considering the objectives of (I) to identify the key parameters of an international mega container port, (II) to compare the five selected container ports (JNPT, Chennai, Singapore, Dubai, and Colombo Ports) according to user of the ports and iii) to measure the growth of selected five container ports’ throughput over time and their comparison. The study is based on both primary and secondary databases. The linear time trend analysis is done to show the trend in quantum of exports, imports and total goods/services handled by individual ports over the years. The comparative trend analysis is done for the selected five ports of cargo traffic handled in terms of Tonnage (weight) and number of containers (TEU’s). The comparative trend analysis is done between containerized and non-containerized cargo traffic in the five selected five ports. The primary data analysis is done comprising of comparative analysis of factor ratings through bar diagrams, statistical inference of factor ratings for the selected five ports, consolidated comparative line charts of factor rating for the selected five ports, consolidated comparative bar charts of factor ratings of the selected five ports and the distribution of ratings (frequency terms). The linear regression model is used to forecast the container capacities required for JNPT Port and Chennai Port by the year 2030. Multiple regression analysis is carried out to measure the impact of selected 34 explanatory variables on the ‘Overall Performance of the Port’ for each of the selected five ports. The research outcome is of high significance to the stakeholders of Indian container handling ports. Indian container port of JNPT and Chennai are benchmarked against international ports such as Singapore, Dubai, and Colombo Ports which are the competing ports in the neighbouring region. The study has analysed the feedback ratings for the selected 35 factors regarding physical infrastructure and services rendered to the port users. This feedback would provide valuable data for carrying out improvements in the facilities provided to the port users. These installations would help the ports’ users to carry out their work in more efficient manner.

Keywords: throughput, twenty equivalent units, TEUs, cargo traffic, shipping lines, freight forwarders

Procedia PDF Downloads 108

Authors: Serhat Tüzün, Tufan Demirel

Abstract:

Decision Support Systems (DSS) have been investigated by researchers and technologists for more than 35 years. This paper analyses the developments in the architecture and software of these systems, provides a systematic analysis for different Web-based DSS approaches and Intelligent Decision-making Technologies (IDT), with the suggestion for future studies. Decision Support Systems literature begins with building model-oriented DSS in the late 1960s, theory developments in the 1970s, and the implementation of financial planning systems and Group DSS in the early and mid-80s. Then it documents the origins of Executive Information Systems, online analytic processing (OLAP) and Business Intelligence. The implementation of Web-based DSS occurred in the mid-1990s. With the beginning of the new millennia, intelligence is the main focus on DSS studies. Web-based technologies are having a major impact on design, development and implementation processes for all types of DSS. Web technologies are being utilized for the development of DSS tools by leading developers of decision support technologies. Major companies are encouraging its customers to port their DSS applications, such as data mining, customer relationship management (CRM) and OLAP systems, to a web-based environment. Similarly, real-time data fed from manufacturing plants are now helping floor managers make decisions regarding production adjustment to ensure that high-quality products are produced and delivered. Web-based DSS are being employed by organizations as decision aids for employees as well as customers. A common usage of Web-based DSS has been to assist customers configure product and service according to their needs. These systems allow individual customers to design their own products by choosing from a menu of attributes, components, prices and delivery options. The Intelligent Decision-making Technologies (IDT) domain is a fast growing area of research that integrates various aspects of computer science and information systems. This includes intelligent systems, intelligent technology, intelligent agents, artificial intelligence, fuzzy logic, neural networks, machine learning, knowledge discovery, computational intelligence, data science, big data analytics, inference engines, recommender systems or engines, and a variety of related disciplines. Innovative applications that emerge using IDT often have a significant impact on decision-making processes in government, industry, business, and academia in general. This is particularly pronounced in finance, accounting, healthcare, computer networks, real-time safety monitoring and crisis response systems. Similarly, IDT is commonly used in military decision-making systems, security, marketing, stock market prediction, and robotics. Even though lots of research studies have been conducted on Decision Support Systems, a systematic analysis on the subject is still missing. Because of this necessity, this paper has been prepared to search recent articles about the DSS. The literature has been deeply reviewed and by classifying previous studies according to their preferences, taxonomy for DSS has been prepared. With the aid of the taxonomic review and the recent developments over the subject, this study aims to analyze the future trends in decision support systems.

Keywords: decision support systems, intelligent decision-making, systematic analysis, taxonomic review

Procedia PDF Downloads 244

Authors: Meiling Yu, Nadia Rouatbi, Khuloud T. Al-Jamal

Abstract:

Background: Treatment of neurological disorders with modern medical and surgical approaches remains difficult. Gene therapy, allowing the delivery of genetic materials that encodes potential therapeutic molecules, represents an attractive option. The treatment of brain diseases with gene therapy requires the gene-editing tool to be delivered efficiently to the central nervous system. In this study, we explored the efficiency of different delivery routes, namely intravenous (i.v.), intra-cranial (i.c.), and intra-nasal (i.n.), to deliver stable nucleic acid-lipid particles (SNALPs) containing gene-editing tools namely Cas9 mRNA and sgRNA encoding for GFP as a reporter protein. We hypothesise that SNALPs can reach the brain and perform gene-editing to different extents depending on the administration route. Intranasal administration (i.n.) offers an attractive and non-invasive way to access the brain circumventing the blood–brain barrier. Successful delivery of gene-editing tools to the brain offers a great opportunity for therapeutic target validation and nucleic acids therapeutics delivery to improve treatment options for a range of neurodegenerative diseases. In this study, we utilised Rosa26-Cas9 knock-in mice, expressing GFP, to study brain distribution and gene-editing efficiency of SNALPs after i.v.; i.c. and i.n. routes of administration. Methods: Single guide RNA (sgRNA) against GFP has been designed and validated by in vitro nuclease assay. SNALPs were formulated and characterised using dynamic light scattering. The encapsulation efficiency of nucleic acids (NA) was measured by RiboGreen™ assay. SNALPs were incubated in serum to assess their ability to protect NA from degradation. Rosa26-Cas9 knock-in mice were i.v., i.n., or i.c. administered with SNALPs to test in vivo gene-editing (GFP knockout) efficiency. SNALPs were given as three doses of 0.64 mg/kg sgGFP following i.v. and i.n. or a single dose of 0.25 mg/kg sgGFP following i.c.. knockout efficiency was assessed after seven days using Sanger Sequencing and Inference of CRISPR Edits (ICE) analysis. In vivo, the biodistribution of DiR labelled SNALPs (SNALPs-DiR) was assessed at 24h post-administration using IVIS Lumina Series III. Results: Serum-stable SNALPs produced were 130-140 nm in diameter with ~90% nucleic acid loading efficiency. SNALPs could reach and stay in the brain for up to 24h following i.v.; i.n. and i.c. administration. Decreasing GFP expression (around 50% after i.v. and i.c. and 20% following i.n.) was confirmed by optical imaging. Despite the small number of mice used, ICE analysis confirmed GFP knockout in mice brains. Additional studies are currently taking place to increase mice numbers. Conclusion: Results confirmed efficient gene knockout achieved by SNALPs in Rosa26-Cas9 knock-in mice expressing GFP following different routes of administrations in the following order i.v.= i.c.> i.n. Each of the administration routes has its pros and cons. The next stages of the project involve assessing gene-editing efficiency in wild-type mice and replacing GFP as a model target with therapeutic target genes implicated in Motor Neuron Disease pathology.

Keywords: CRISPR, nanoparticles, brain diseases, administration routes

Procedia PDF Downloads 62

Authors: Partha Saha, Uttam Kumar Banerjee

Abstract:

Maternal and Child Healthcare (MCH) has always been regarded as one of the important issues globally. Reduction of maternal and child mortality rates and increase of healthcare service coverage were declared as one of the targets in Millennium Development Goals till 2015 and thereafter as an important component of the Sustainable Development Goals. Over the last decade, worldwide MCH indicators have improved but could not match the expected levels. Progress of both maternal and child mortality rates have been monitored by several researchers. Each of the studies has stated that only less than 26% of low-income and middle income countries (LMICs) were on track to achieve targets as prescribed by MDG4. Average worldwide annual rate of reduction of under-five mortality rate and maternal mortality rate were 2.2% and 1.9% as on 2011 respectively whereas rates should be minimum 4.4% and 5.5% annually to achieve targets. In spite of having proven healthcare interventions for both mothers and children, those could not be scaled up to the required volume due to fragmented health systems, especially in the developing and under-developed countries. In this research, a knowledge discovery based interactive Decision Support System (DSS) has been developed on web platform which would assist healthcare policy makers to develop evidence-based policies. To achieve desirable results in MCH, efficient resource planning is very much required. In maximum LMICs, resources are big constraint. Knowledge, generated through this system, would help healthcare managers to develop strategic resource planning for combatting with issues like huge inequity and less coverage in MCH. This system would help healthcare managers to accomplish following four tasks. Those are a) comprehending region wise conditions of variables related with MCH, b) identifying relationships within variables, c) segmenting regions based on variables status, and d) finding out segment wise key influential variables which have major impact on healthcare indicators. Whole system development process has been divided into three phases. Those were i) identifying contemporary issues related with MCH services and policy making; ii) development of the system; and iii) verification and validation of the system. More than 90 variables under three categories, such as a) educational, social, and economic parameters; b) MCH interventions; and c) health system building blocks have been included into this web-based DSS and five separate modules have been developed under the system. First module has been designed for analysing current healthcare scenario. Second module would help healthcare managers to understand correlations among variables. Third module would reveal frequently-occurring incidents along with different MCH interventions. Fourth module would segment regions based on previously mentioned three categories and in fifth module, segment-wise key influential interventions will be identified. India has been considered as case study area in this research. Data of 601 districts of India has been used for inspecting effectiveness of those developed modules. This system has been developed by importing different statistical and data mining techniques on Web platform. Policy makers would be able to generate different scenarios from the system before drawing any inference, aided by its interactive capability.

Keywords: maternal and child heathcare, decision support systems, data mining techniques, low and middle income countries

Procedia PDF Downloads 219

Authors: Chung Yan Natalia Yeung, Man Kit Indy Ho, Kin Yu Stan Chan, Ho Pui Kipper Lam, Man Wah Genie Tong, Tze Chung Jim Luk

Abstract:

Excessive foot pronation is a well-known risk factor of knee and foot injuries such as patellofemoral pain, patellar and Achilles tendinopathy, and plantar fasciitis. Low-Dye taping (LDT) application is not uncommon for basketball players to control excessive foot pronation for pain control and injury prevention. The primary potential benefits of using LDT include providing additional supports to medial longitudinal arch and restricting the excessive midfoot and subtalar motion in weight-bearing activities such as running and landing. Meanwhile, restrictions provided by the rigid tape may also potentially limit functional joint movements and sports performance. Coaches and athletes need to weigh the potential benefits and harmful effects before making a decision if applying LDT technique is worthwhile or not. However, the influence of using LDT on basketball-related performance such as explosive and reactive strength is not well understood. Therefore, the purpose of this study was to investigate the change of drop jump (DJ) and countermovement jump (CMJ) performance before and after LDT application for collegiate male basketball players. In this within-subject crossover study, 12 healthy male basketball players (age: 21.7 ± 2.5 years) with at least 3-year regular basketball training experience were recruited. Navicular drop (ND) test was adopted as the screening and only those with excessive pronation (ND ≥ 10mm) were included. Participants with recent lower limb injury history were excluded. Recruited subjects were required to perform both ND, DJ (on a platform of 40cm height) and CMJ (without arms swing) tests in series during taped and non-taped conditions in the counterbalanced order. Reactive strength index (RSI) was calculated by using the flight time divided by the ground contact time measured. For DJ and CMJ tests, the best of three trials was used for analysis. The difference between taped and non-taped conditions for each test was further calculated through standardized effect ± 90% confidence intervals (CI) with clinical magnitude-based inference (MBI). Paired samples T-test showed significant decrease in ND (-4.68 ± 1.44mm; 95% CI: -3.77, -5.60; p < 0.05) while MBI demonstrated most likely beneficial and large effect (standardize effect: -1.59 ± 0.27) in LDT condition. For DJ test, significant increase in both flight time (25.25 ± 29.96ms; 95% CI: 6.22, 44.28; p < 0.05) and RSI (0.22 ± 0.22; 95% CI: 0.08, 0.36; p < 0.05) were observed. In taped condition, MBI showed very likely beneficial and moderate effect (standardized effect: 0.77 ± 0.49) in flight time, possibly beneficial and small effect (standardized effect: -0.26 ± 0.29) in ground contact time and very likely beneficial and moderate effect (standardized effect: 0.77 ± 0.42) in RSI. No significant difference in CMJ was observed (95% CI: -2.73, 2.08; p > 0.05). For basketball players with pes planus, applying LDT could substantially support the foot by elevating the navicular height and potentially provide acute beneficial effects in reactive strength performance. Meanwhile, no significant harmful effect on CMJ was observed. Basketball players may consider applying LDT before the game or training to enhance the reactive strength performance. However since the observed effects in this study could not generalize to other players without excessive foot pronation, further studies on players with normal foot arch or navicular height are recommended.

Keywords: flight time, pes planus, pronated foot, reactive strength index

Procedia PDF Downloads 127

Authors: André Python

Abstract:

To this day, terrorism persists as a worldwide threat, exemplified by the recent deadly attacks in January 2015 in Paris and the ongoing massacres perpetrated by ISIS in Iraq and Syria. In response to this threat, states deploy various counterterrorism measures, the cost of which could be reduced through effective preventive measures. In order to increase the efficiency of preventive measures, policy-makers may benefit from accurate predictive models that are able to capture the complex spatial dynamics of terrorism occurring at a local scale. Despite empirical research carried out at country-level that has confirmed theories explaining the diffusion processes of terrorism across space and time, scholars have failed to assess diffusion’s theories on a local scale. Moreover, since scholars have not made the most of recent statistical modelling approaches, they have been unable to build up predictive models accurate in both space and time. In an effort to address these shortcomings, this research suggests a novel approach to systematically assess the theories of terrorism’s diffusion on a local scale and provide a predictive model of the local spatial dynamics of terrorism worldwide. With a focus on the lethal terrorist events that occurred after 9/11, this paper addresses the following question: why and how does lethal terrorism diffuse in space and time? Based on geolocalised data on worldwide terrorist attacks and covariates gathered from 2002 to 2013, a binomial spatio-temporal point process is used to model the probability of terrorist attacks on a sphere (the world), the surface of which is discretised in the form of Delaunay triangles and refined in areas of specific interest. Within a Bayesian framework, the model is fitted through an integrated nested Laplace approximation - a recent fitting approach that computes fast and accurate estimates of posterior marginals. Hence, for each location in the world, the model provides a probability of encountering a lethal terrorist attack and measures of volatility, which inform on the model’s predictability. Diffusion processes are visualised through interactive maps that highlight space-time variations in the probability and volatility of encountering a lethal attack from 2002 to 2013. Based on the previous twelve years of observation, the location and lethality of terrorist events in 2014 are statistically accurately predicted. Throughout the global scope of this research, local diffusion processes such as escalation and relocation are systematically examined: the former process describes an expansion from high concentration areas of lethal terrorist events (hotspots) to neighbouring areas, while the latter is characterised by changes in the location of hotspots. By controlling for the effect of geographical, economical and demographic variables, the results of the model suggest that the diffusion processes of lethal terrorism are jointly driven by contagious and non-contagious factors that operate on a local scale – as predicted by theories of diffusion. Moreover, by providing a quantitative measure of predictability, the model prevents policy-makers from making decisions based on highly uncertain predictions. Ultimately, this research may provide important complementary tools to enhance the efficiency of policies that aim to prevent and combat terrorism.

Keywords: diffusion process, terrorism, spatial dynamics, spatio-temporal modeling

Procedia PDF Downloads 313

Authors: Alexander David Natanson

Abstract:

This paper focuses on the effect of Sanctuary and Immigration and Customs Enforcement (ICE) policies on local economies. "Sanctuary cities" refers to municipal jurisdictions that limit their cooperation with the federal government's efforts to enforce immigration. Using county-level data from the American Community Survey and ICE data on economic indicators from 2006 to 2018, this study isolates the effects of local immigration policies on U.S. counties. The investigation is accomplished by simultaneously studying the policies' effects in counties where immigrants' families are persecuted via collaboration with Immigration and Customs Enforcement (ICE), in contrast to counties that provide protections. The analysis includes a difference-in-difference & two-way fixed effect model. Results are robust to nearest-neighbor matching, after the random assignment of treatment, after running estimations using different cutoffs for immigration policies, and with a regression discontinuity model comparing bordering counties with opposite policies. Results are also robust after restricting the data to a single-year policy adoption, using the Sun and Abraham estimator, and with event-study estimation to deal with the staggered treatment issue. In addition, the study reverses the estimation to understand what drives the decision to choose policies to detect the presence of reverse causality biases in the estimated policy impact on economic factors. The evidence demonstrates that providing protections to undocumented immigrants increases economic activity. The estimates show gains in per capita income ranging from 3.1 to 7.2, median wages between 1.7 to 2.6, and GDP between 2.4 to 4.1 percent. Regarding labor, sanctuary counties saw increases in total employment between 2.3 to 4 percent, and the unemployment rate declined from 12 to 17 percent. The data further shows that ICE policies have no statistically significant effects on income, median wages, or GDP but adverse effects on total employment, with declines from 1 to 2 percent, mostly in rural counties, and an increase in unemployment of around 7 percent in urban counties. In addition, results show a decline in the foreign-born population in ICE counties but no changes in sanctuary counties. The study also finds similar results for sanctuary counties when separating the data between urban, rural, educational attainment, gender, ethnic groups, economic quintiles, and the number of business establishments. The takeaway from this study is that institutional inclusion creates the dynamic nature of an economy, as inclusion allows for economic expansion due to the extension of fundamental freedoms to newcomers. Inclusive policies show positive effects on economic outcomes with no evident increase in population. To make sense of these results, the hypothesis and theoretical model propose that inclusive immigration policies play an essential role in conditioning the effect of immigration by decreasing uncertainties and constraints for immigrants' interaction in their communities, decreasing the cost from fear of deportation or the constant fear of criminalization and optimize their human capital.

Keywords: inclusive and exclusive institutions, post matching, fixed effect, time trend, regression discontinuity, difference-in-difference, randomization inference and sun, Abraham estimator

Procedia PDF Downloads 56

Authors: Daniel Puciato, Piotr Oleśniewicz, Julita Markiewicz-Patkowska, Krzysztof Widawski, Michał Rozpara, Władysław Mynarski, Agnieszka Gawlik, Małgorzata Dębska, Soňa Jandová

Abstract:

Physical activity as a part of people’s everyday life reduces the risk of many diseases, including those induced by lifestyle, e.g. obesity, type 2 diabetes, osteoporosis, coronary heart disease, degenerative arthritis, and certain types of cancer. That refers particularly to professionally active people, including the early senior group working on non-manual positions. The aim of the study is to evaluate the relationship between physical activity and the socio-economic status of non-manual workers from Wroclaw—one of the biggest cities in Poland, a model setting for such investigations in this part of Europe. The crucial problem in the research is to find out the percentage of respondents who meet the health-related recommendations of the World Health Organization (WHO) concerning the volume, frequency, and intensity of physical activity, as well as to establish if the most important socio-economic factors, such as gender, age, education, marital status, per capita income, savings and debt, determine the compliance with the WHO physical activity recommendations. During the research, conducted in 2013, 1,170 people (611 women and 559 men) aged 21–60 years were examined. A diagnostic poll method was applied to collect the data. Physical activity was measured with the use of the short form of the International Physical Activity Questionnaire with extended socio-demographic questions, i.e. concerning gender, age, education, marital status, income, savings or debts. To evaluate the relationship between physical activity and selected socio-economic factors, logistic regression was used (odds ratio statistics). Statistical inference was conducted on the adopted ex ante probability level of p<0.05. The majority of respondents met the volume of physical effort recommended for health benefits. It was particularly noticeable in the case of the examined men. The probability of compliance with the WHO physical activity recommendations was highest for workers aged 21–30 years with secondary or higher education who were single, received highest incomes and had savings. The results indicate the relations between physical activity and socio-economic status in the examined women and men. People with lower socio-economic status (e.g. manual workers) are physically active primarily at work, whereas those better educated and wealthier implement physical effort primarily in their leisure time. Among the investigated subjects, the youngest group of non-manual workers have the best chances to meet the WHO standards of physical activity. The study also confirms that secondary education has a positive effect on the public awareness on the role of physical activity in human life. In general, the analysis of the research indicates that there is a relationship between physical activity and some socio-economic factors of the respondents, such as gender, age, education, marital status, income per capita, and the possession of savings. Although the obtained results cannot be applied for the general population, they show some important trends that will be verified in subsequent studies conducted by the authors of the paper.

Keywords: IPAQ, nonmanual workers, physical activity, socioeconomic factors, WHO

Procedia PDF Downloads 496

Authors: Pei Yi Lin

Abstract:

Objective: The objective of this study is to use machine learning methods to build an early prediction classifier model for acute delirium to improve the quality of medical care for intensive care patients. Background: Delirium is a common acute and sudden disturbance of consciousness in critically ill patients. After the occurrence, it is easy to prolong the length of hospital stay and increase medical costs and mortality. In 2021, the incidence of delirium in the intensive care unit of internal medicine was as high as 59.78%, which indirectly prolonged the average length of hospital stay by 8.28 days, and the mortality rate is about 2.22% in the past three years. Therefore, it is expected to build a delirium prediction classifier through big data analysis and machine learning methods to detect delirium early. Method: This study is a retrospective study, using the artificial intelligence big data database to extract the characteristic factors related to delirium in intensive care unit patients and let the machine learn. The study included patients aged over 20 years old who were admitted to the intensive care unit between May 1, 2022, and December 31, 2022, excluding GCS assessment <4 points, admission to ICU for less than 24 hours, and CAM-ICU evaluation. The CAMICU delirium assessment results every 8 hours within 30 days of hospitalization are regarded as an event, and the cumulative data from ICU admission to the prediction time point are extracted to predict the possibility of delirium occurring in the next 8 hours, and collect a total of 63,754 research case data, extract 12 feature selections to train the model, including age, sex, average ICU stay hours, visual and auditory abnormalities, RASS assessment score, APACHE-II Score score, number of invasive catheters indwelling, restraint and sedative and hypnotic drugs. Through feature data cleaning, processing and KNN interpolation method supplementation, a total of 54595 research case events were extracted to provide machine learning model analysis, using the research events from May 01 to November 30, 2022, as the model training data, 80% of which is the training set for model training, and 20% for the internal verification of the verification set, and then from December 01 to December 2022 The CU research event on the 31st is an external verification set data, and finally the model inference and performance evaluation are performed, and then the model has trained again by adjusting the model parameters. Results: In this study, XG Boost, Random Forest, Logistic Regression, and Decision Tree were used to analyze and compare four machine learning models. The average accuracy rate of internal verification was highest in Random Forest (AUC=0.86), and the average accuracy rate of external verification was in Random Forest and XG Boost was the highest, AUC was 0.86, and the average accuracy of cross-validation was the highest in Random Forest (ACC=0.77). Conclusion: Clinically, medical staff usually conduct CAM-ICU assessments at the bedside of critically ill patients in clinical practice, but there is a lack of machine learning classification methods to assist ICU patients in real-time assessment, resulting in the inability to provide more objective and continuous monitoring data to assist Clinical staff can more accurately identify and predict the occurrence of delirium in patients. It is hoped that the development and construction of predictive models through machine learning can predict delirium early and immediately, make clinical decisions at the best time, and cooperate with PADIS delirium care measures to provide individualized non-drug interventional care measures to maintain patient safety, and then Improve the quality of care.

Keywords: critically ill patients, machine learning methods, delirium prediction, classifier model

Procedia PDF Downloads 31